Abstract
Geographic phenomena exist within a multi-dimensional space–time continuum. Dynamic geographic phenomena at all levels of scale can be conceptualized and represented as spatiotemporal patterns, space–time processes, or events—changes within or between objects that are experienced as bounded by psychologically discreet beginnings and ends. Humans rarely care about spatiotemporal entities in isolation. Visualization and analysis approaches that focus on individual spatiotemporal phenomena in isolation are likely doomed to failure because they miss the relational structure humans use to process and reason about events. We contend that a static and geometric decompositional approach to spatiotemporal patterns and processes limits the tools that can be applied to a broad class of spatiotemporal data that are important to users. This class includes events where there is a spatiotemporal coordination among or within objects, such as a car changing its movement direction because of an approaching car, or a hurricane not making landfall because of changing atmospheric conditions. Often such coordination allows inferences about causal relations among the components of an event. In this chapter we argue for the need for perceptually salient and cognitively inspired animated displays that help humans more effectively and efficiently detect relationships in complex events.